Description
Imagine being at a social gathering, attempting to capture a memorable moment, only to realize later that a still picture falls short of truly encapsulating the experience. With Apple's introduction of Live images on iPhones, modern mobile photography has evolved, adding interactivity and captivation to moments captured. Unlike still pictures, Live images not only freeze a moment but also infuse it with audio and subtle body language nuances, akin to the moving pictures concept depicted in Harry Potter movies. This feature brings an essence of human interaction to life within a single image, enriching the viewing experience.
Moreover, Live images extend beyond mere moments; they offer potential insights into human behavior and group dynamics, making them valuable for both research and industry applications. Despite their rising popularity and research potential, comprehensive live image datasets have been notably absent in the literature.
Here, we present an extensive, novel, large-scale, self-curated Live image dataset featuring diverse groups of human subjects, both with and without face masks. The dataset comprises 17,850 live images saved as MOV files, accompanied by their respective Key Photos saved in HEIC format. This dataset is made publicly available solely for research purposes, with users expected to cite the following paper and other related papers that stem from this work.
O. Kulkarni, A. Mishra, S. Arora, V. K. Singh, and P. K. Atrey. LivePics-24: A Multi-person Multi-camera Multi-settings Live Photos Dataset. MIPR'24: IEEE International Conference on Multimedia Information Processing ad Retrieval, San Jose, CA, USA, August 2024.